Background/ Rationale: Falls are a significant problem in the older population and are responsible for nearly half of all injury-related deaths in persons over the age of 65. Many clinical screening tests can predict fall risk, but none of these tests incorporate or assess responses to gait perturbations that are likely to initiate most falls. A history of falls is one o the best predictors of future falls in clinical practice and is the gold-standard outcome for research studies of fall- reduction interventions. However, self-report may underestimate actual falls and does not account for exposure to risk, which could have substantial effects on the likelihood of falling. Furthermore, issues with self-report and the relative infrequency of falls necessitates large sample sizes to detect significant reductions in fall risk from interventions. Following thes large samples over a long period of time results in high costs per study and thus fewer studies that can be funded to evaluate fall-reduction interventions. Falls have been safely induced in controlled laboratory settings using a single perturbation at a time, but subjects inevitably adapt their gait pattern to minimize their risk from the particular perturbation used. The specific adaptations vary by perturbation type, but all are departures from habitual gait patterns that reduce the ecological validity of the results. The possibility of various types of gait perturbatio during a single testing session could reduce or eliminate these adaptations and thus improve the ecological validity of laboratory-induced fall recovery studies. Objectives- The objectives of this three-year, laboratory-based study are as follows: Development: Design and construct active and passive gait perturbation systems. Testing: Verify safety, feasibility, and efficacy of active and passive gait perturbation systems and testing protocols and then complete human subjects testing. Analyisis: Complete data processing and analysis, test hypotheses, and formulate new hypotheses for future research. Methods: This developmental laboratory-based study will be completed in three phases: The Development Phase will consist of the design, construction, debugging, and iteration of the active and passive gait perturbation systems. The active surface will consist of five discrete perturbation systems embedded into the laboratory flooring. This will evaluate the subject's ability to recover from impending falls. The passive surface will be a 4m x 0.5m walkway consisting of small wooden obstacles in a layer of soft foam sandwiched between two layers of low-pile carpet. This will assess the subject's ability to maintain a consistent gait pattern and speed when subjected to small unseen deviations in the walkway surface. Only when safety (a safety harness will be worn at all times) and effectiveness of these systems have been satisfactorily established will human subjects testing begin. The Testing Phase will begin by testing 10 unimpaired young subjects to fine-tune procedures and verify safety, then move on to evaluating 60 older subjects- 30 non-fallers (not fallen in last yea) and 30 fallers (fallen within the last 6 months).. The Analysis Phase of will consist of will consit of biomechanical data analysis using Visual3D and statistical analysis using SAS.